Replication Materials for "When Corporations Are People:  Agent Talk and the Development of Organizational Actorhood, 1890-1934"
by Carly Knight (carly.knight@nyu.edu), July 31, 20222

***

The files that accompany this readme allow the researcher to replicate all figures and tables in the SMR publication above.

The original data (and full text) of the articles come from Proquest's Historical Newspapers Text-as-Data Collection (https://guides.nyu.edu/tdm/proquest-historical-newspapers).


There are three main datasets used in the analysis:

(1) agency_files.Rdata: File-level (435,655 rows; 29 columns)
This is the main dataset used it in the analysis. It contains file-level variables associated with the 435,655 articles in the analysis. 
This file is used to generate Figure 1, Table 3, Figure 5, and Figure 6. 

(2) Agency_entities.Rdata: Entity-level (868,183 rows; 13 columns)
This dataset contains entity level, extracted features from the articles in agency_files.Rdata. Per NYU's purchase license agreement, I am able to provide an extracted features dataset that lists two-token co-occurences along with the metadata about the text in which it appears.  This file is used to generate Table 4 and Figure 3. 

(3) stm12.Rdata: STM output file
This is used to generate Figure 4. 

****

The main replication file for the tables and figures that appear in the paper is "Figures and Tables for When Corporations Are People.Rmd"

Replication materials for analyzes performed in the online supplement are available from the author on request.
